def test_save_parameter_variation():
print('>>> test_save_parameter_variation')
db = H5DB('parameter_variation_test.h5', [ParameterVariation, KeyValuePair])
db.open()
step_size = KeyValuePair(key='value', value=60)
#step_size.key = 'value'
#step_size.value = 60
eta_pc = KeyValuePair()
eta_pc.key = 'value'
eta_pc.value = [-2.109566, 0.403556, 97.110770]
parameter_variation = ParameterVariation()
parameter_variation.parameter_name = "param_name"
parameter_variation.variation_mode = "constant"
parameter_variation.variation_arguments = [step_size]
db.save_object(parameter_variation)
parameter_variation = ParameterVariation()
parameter_variation.parameter_name = "param_name"
parameter_variation.variation_mode = "constant"
parameter_variation.variation_arguments = [eta_pc]
db.save_object(parameter_variation)
load_and_print_objects(db, ParameterVariation)
db.close()
#print(yaml.dump(parameter_variation))
print('<<< test_save_parameter_variation')
def test_save_key_value_pair():
db = H5DB('keyvaluepair_test.h5', [KeyValuePair])
db.open()
kvp = KeyValuePair()
kvp.key = '123'
kvp.value = 123
db.save_object(kvp)
load_and_print_objects(db, KeyValuePair)
db.close()
def test_save_sim_config():
db = H5DB('sim_configs.h5', [SimConfig, KeyValuePair])
db.open()
obj = SimConfig()
obj.arguments = [
KeyValuePair(key='python', value='234'),
KeyValuePair(key='cwd', value='.')]
db.save_object(obj)
load_and_print_objects(db, SimConfig)
#print(yaml.dump(obj))
db.close()
def test_save_strategy_config():
print('>>> test_save_strategy_config')
db = H5DB('strategy_test.h5', [StrategyConfig, KeyValuePair])
db.open()
parameters = []
for i in range(5):
parameters.append(KeyValuePair())
parameters[i].key = 'param_%d' % (i)
parameters[i].value = i
strategy = StrategyConfig()
strategy.name = "examples strategie"
strategy.arguments = parameters
db.save_object(strategy)
load_and_print_objects(db, StrategyConfig)
db.close()
#print(yaml.dump(strategy))
print('<<< test_save_strategy_config')
def battery_sim():
print('>>> battery_sim')
db = H5DB('batterysimtest.h5', [SimConfig, ModelStructure, VirtualState, SamplerConfig, ParameterVariation,
StrategyConfig, KeyValuePair, SurrogateModel, ApproximationFunctionConfig])
db.open()
# sim config
sim_config = SimConfig()
sim_config.arguments = [
KeyValuePair(key='python', value='pysimmods.batterysim.battery_mosaik:BatterySimulator'),
KeyValuePair(key='cwd', value='.')]
# model structure
model_structure = ModelStructure()
model_structure.simulator_parameters = ['step_size']
model_structure.model_parameters = ['capacity', 'P_el_max', 'P_el_min', 'eta_pc', 'SoC']
model_structure.model_inputs = ['P_el_set']
model_structure.model_outputs = ['P_el', 'SoC']
soc = VirtualState(name='SoC', init_attribute='SoC', update_attribute='SoC')
model_structure.virtual_states = [soc]
# sampling strategy
num_samples = KeyValuePair(key='num_samples', value=1000)
strategy = StrategyConfig(name='lhs')
strategy.arguments = [num_samples]
# parameter variations
step_size_variation = ParameterVariation(parameter_name='step_size', variation_mode='constant')
step_size_variation.variation_arguments = [KeyValuePair(key='value', value=60)]
capacity_variation = ParameterVariation(parameter_name='capacity', variation_mode='constant')
capacity_variation.variation_arguments = [KeyValuePair(key='value', value=5.0)]
p_el_max_variation = ParameterVariation(parameter_name='P_el_max', variation_mode='constant')
p_el_max_variation.variation_arguments = [KeyValuePair(key='value', value=1000.0)]
p_el_min_variation = ParameterVariation(parameter_name='P_el_min', variation_mode='constant')
p_el_min_variation.variation_arguments = [KeyValuePair(key='value', value=-1000.0)]
eta_pc_variation = ParameterVariation(parameter_name='eta_pc', variation_mode='constant')
eta_pc_variation.variation_arguments = [KeyValuePair(key='value', value=[-2.109566, 0.403556, 97.110770])]
p_el_set_variation = ParameterVariation(parameter_name='P_el_set', variation_mode='range_of_real_numbers')
p_el_set_variation.variation_arguments = [KeyValuePair(key='min', value=-1000), KeyValuePair(key='max', value=1000)]
soc_variation = ParameterVariation(parameter_name='SoC', variation_mode='range_of_real_numbers')
soc_variation.variation_arguments = [KeyValuePair(key='min', value=0.0), KeyValuePair(key='max', value=1.0)]
parameter_variations = [step_size_variation, capacity_variation, p_el_max_variation, p_el_min_variation,
eta_pc_variation, p_el_set_variation, soc_variation]
# sampler configuration
sampler_config = SamplerConfig(name='LHS_N_100')
sampler_config.sim_config = sim_config
sampler_config.strategy = strategy
sampler_config.model_structure = model_structure
sampler_config.parameter_variations = parameter_variations
#db.save_object(sampler_config)
# trainer configuration
approx = ApproximationFunctionConfig()
approx.inputs = ['P_el_set', 'SoC']
approx.outputs = ['P_el', 'SoC']
approx.model_type = 'Kriging'
approx.model_arguments = []
approx.trainer_options = []
sm = SurrogateModel()
sm.name = 'mySurrogateModel'
sm.approximation_functions = [approx]
sm.sampler_configuration = sampler_config
db.save_object(sm)
sm2 = SurrogateModel()
sm2.name = 'mySurrogateModel'
sm2.approximation_functions = [approx]
sm2.sampler_configuration = sampler_config
db.save_object(sm2)
load_and_print_objects(db, SurrogateModel)
#self.inputs = List
#self.outputs = List
#self.model_type = Scalar
#self.model_arguments = ObjectList
#self.trainer_options = ObjectList
db.close()
#print(yaml.dump(sm))
print('>>> battery_sim')